The increase in purchasing alternatives forces companies to focus on the factors affecting consumer buying behavior and to determine what they are paying attention to. This study aims to research consumers’ numerical and sensory information preferences and to determine what information the consumers take most on the packaging. In this study, data obtained from the participants’ have been studied with machine learning methods according to the information type they prefer. The data of the study were collected using a questionnaire form. Sample selection was done by using a convenience sampling method. In this study, consumers are classified with 18 big data-based machine learning methods. According to the results, the Linear Support Vector Machine (LSVM) method supports the best result. The use of big data analysis methods in marketing is the contribution of this study to the current literature. In practice, determining the type of information needs of the target consumers helps package designers in designing the information to be included in the packaging.